Anthropic's 123-library experiment shows that LLMs can scale proof writing. A deployment assurance manifest shows what was proved, what ships, and wha